== Physical Plan ==
VeloxColumnarToRow (36)
+- TakeOrderedAndProjectExecTransformer (35)
   +- ^ FilterExecTransformer (33)
      +- ^ RegularHashAggregateExecTransformer (32)
         +- ^ InputIteratorTransformer (31)
            +- ColumnarExchange (29)
               +- VeloxResizeBatches (28)
                  +- ^ ProjectExecTransformer (26)
                     +- ^ FlushableHashAggregateExecTransformer (25)
                        +- ^ ProjectExecTransformer (24)
                           +- ^ BroadcastHashJoinExecTransformer Inner BuildRight (23)
                              :- ^ ProjectExecTransformer (16)
                              :  +- ^ BroadcastHashJoinExecTransformer Inner BuildRight (15)
                              :     :- ^ ProjectExecTransformer (11)
                              :     :  +- ^ BroadcastHashJoinExecTransformer Inner BuildRight (10)
                              :     :     :- ^ FilterExecTransformer (2)
                              :     :     :  +- ^ FileSourceScanExecTransformer parquet spark_catalog.default.inventory (1)
                              :     :     +- ^ InputIteratorTransformer (9)
                              :     :        +- ColumnarBroadcastExchange (7)
                              :     :           +- ^ ProjectExecTransformer (5)
                              :     :              +- ^ FilterExecTransformer (4)
                              :     :                 +- ^ FileSourceScanExecTransformer parquet spark_catalog.default.item (3)
                              :     +- ^ InputIteratorTransformer (14)
                              :        +- ReusedExchange (12)
                              +- ^ InputIteratorTransformer (22)
                                 +- ColumnarBroadcastExchange (20)
                                    +- ^ FilterExecTransformer (18)
                                       +- ^ FileSourceScanExecTransformer parquet spark_catalog.default.warehouse (17)


(1) FileSourceScanExecTransformer parquet spark_catalog.default.inventory
Output [4]: [inv_item_sk#1, inv_warehouse_sk#2, inv_quantity_on_hand#3, inv_date_sk#4]
Batched: true
Location: InMemoryFileIndex []
PartitionFilters: [isnotnull(inv_date_sk#4), dynamicpruningexpression(inv_date_sk#4 IN dynamicpruning#5)]
PushedFilters: [IsNotNull(inv_warehouse_sk), IsNotNull(inv_item_sk)]
ReadSchema: struct<inv_item_sk:int,inv_warehouse_sk:int,inv_quantity_on_hand:int>

(2) FilterExecTransformer
Input [4]: [inv_item_sk#1, inv_warehouse_sk#2, inv_quantity_on_hand#3, inv_date_sk#4]
Arguments: (isnotnull(inv_warehouse_sk#2) AND isnotnull(inv_item_sk#1))

(3) FileSourceScanExecTransformer parquet spark_catalog.default.item
Output [3]: [i_item_sk#6, i_item_id#7, i_current_price#8]
Batched: true
Location: InMemoryFileIndex [{warehouse_dir}/item]
PushedFilters: [IsNotNull(i_current_price), GreaterThanOrEqual(i_current_price,0.99), LessThanOrEqual(i_current_price,1.49), IsNotNull(i_item_sk)]
ReadSchema: struct<i_item_sk:int,i_item_id:string,i_current_price:decimal(7,2)>

(4) FilterExecTransformer
Input [3]: [i_item_sk#6, i_item_id#7, i_current_price#8]
Arguments: (((isnotnull(i_current_price#8) AND (i_current_price#8 >= 0.99)) AND (i_current_price#8 <= 1.49)) AND isnotnull(i_item_sk#6))

(5) ProjectExecTransformer
Output [2]: [i_item_sk#6, i_item_id#7]
Input [3]: [i_item_sk#6, i_item_id#7, i_current_price#8]

(6) WholeStageCodegenTransformer (2)
Input [2]: [i_item_sk#6, i_item_id#7]
Arguments: false

(7) ColumnarBroadcastExchange
Input [2]: [i_item_sk#6, i_item_id#7]
Arguments: HashedRelationBroadcastMode(List(cast(input[0, int, true] as bigint)),false), [plan_id=1]

(8) InputAdapter
Input [2]: [i_item_sk#6, i_item_id#7]

(9) InputIteratorTransformer
Input [2]: [i_item_sk#6, i_item_id#7]

(10) BroadcastHashJoinExecTransformer
Left keys [1]: [inv_item_sk#1]
Right keys [1]: [i_item_sk#6]
Join type: Inner
Join condition: None

(11) ProjectExecTransformer
Output [4]: [inv_warehouse_sk#2, inv_quantity_on_hand#3, inv_date_sk#4, i_item_id#7]
Input [6]: [inv_item_sk#1, inv_warehouse_sk#2, inv_quantity_on_hand#3, inv_date_sk#4, i_item_sk#6, i_item_id#7]

(12) ReusedExchange [Reuses operator id: 40]
Output [2]: [d_date_sk#9, d_date#10]

(13) InputAdapter
Input [2]: [d_date_sk#9, d_date#10]

(14) InputIteratorTransformer
Input [2]: [d_date_sk#9, d_date#10]

(15) BroadcastHashJoinExecTransformer
Left keys [1]: [inv_date_sk#4]
Right keys [1]: [d_date_sk#9]
Join type: Inner
Join condition: None

(16) ProjectExecTransformer
Output [4]: [inv_warehouse_sk#2, inv_quantity_on_hand#3, i_item_id#7, d_date#10]
Input [6]: [inv_warehouse_sk#2, inv_quantity_on_hand#3, inv_date_sk#4, i_item_id#7, d_date_sk#9, d_date#10]

(17) FileSourceScanExecTransformer parquet spark_catalog.default.warehouse
Output [2]: [w_warehouse_sk#11, w_warehouse_name#12]
Batched: true
Location: InMemoryFileIndex [{warehouse_dir}/warehouse]
PushedFilters: [IsNotNull(w_warehouse_sk)]
ReadSchema: struct<w_warehouse_sk:int,w_warehouse_name:string>

(18) FilterExecTransformer
Input [2]: [w_warehouse_sk#11, w_warehouse_name#12]
Arguments: isnotnull(w_warehouse_sk#11)

(19) WholeStageCodegenTransformer (4)
Input [2]: [w_warehouse_sk#11, w_warehouse_name#12]
Arguments: false

(20) ColumnarBroadcastExchange
Input [2]: [w_warehouse_sk#11, w_warehouse_name#12]
Arguments: HashedRelationBroadcastMode(List(cast(input[0, int, false] as bigint)),false), [plan_id=2]

(21) InputAdapter
Input [2]: [w_warehouse_sk#11, w_warehouse_name#12]

(22) InputIteratorTransformer
Input [2]: [w_warehouse_sk#11, w_warehouse_name#12]

(23) BroadcastHashJoinExecTransformer
Left keys [1]: [inv_warehouse_sk#2]
Right keys [1]: [w_warehouse_sk#11]
Join type: Inner
Join condition: None

(24) ProjectExecTransformer
Output [4]: [w_warehouse_name#12, i_item_id#7, CASE WHEN (d_date#10 < 2000-03-11) THEN inv_quantity_on_hand#3 ELSE 0 END AS _pre_1#13, CASE WHEN (d_date#10 >= 2000-03-11) THEN inv_quantity_on_hand#3 ELSE 0 END AS _pre_2#14]
Input [6]: [inv_warehouse_sk#2, inv_quantity_on_hand#3, i_item_id#7, d_date#10, w_warehouse_sk#11, w_warehouse_name#12]

(25) FlushableHashAggregateExecTransformer
Input [4]: [w_warehouse_name#12, i_item_id#7, _pre_1#13, _pre_2#14]
Keys [2]: [w_warehouse_name#12, i_item_id#7]
Functions [2]: [partial_sum(_pre_1#13), partial_sum(_pre_2#14)]
Aggregate Attributes [2]: [sum#15, sum#16]
Results [4]: [w_warehouse_name#12, i_item_id#7, sum#17, sum#18]

(26) ProjectExecTransformer
Output [5]: [hash(w_warehouse_name#12, i_item_id#7, 42) AS hash_partition_key#19, w_warehouse_name#12, i_item_id#7, sum#17, sum#18]
Input [4]: [w_warehouse_name#12, i_item_id#7, sum#17, sum#18]

(27) WholeStageCodegenTransformer (5)
Input [5]: [hash_partition_key#19, w_warehouse_name#12, i_item_id#7, sum#17, sum#18]
Arguments: false

(28) VeloxResizeBatches
Input [5]: [hash_partition_key#19, w_warehouse_name#12, i_item_id#7, sum#17, sum#18]
Arguments: 1024, 2147483647, 10485760

(29) ColumnarExchange
Input [5]: [hash_partition_key#19, w_warehouse_name#12, i_item_id#7, sum#17, sum#18]
Arguments: hashpartitioning(w_warehouse_name#12, i_item_id#7, 1), ENSURE_REQUIREMENTS, [w_warehouse_name#12, i_item_id#7, sum#17, sum#18], [plan_id=3], [shuffle_writer_type=hash]

(30) InputAdapter
Input [4]: [w_warehouse_name#12, i_item_id#7, sum#17, sum#18]

(31) InputIteratorTransformer
Input [4]: [w_warehouse_name#12, i_item_id#7, sum#17, sum#18]

(32) RegularHashAggregateExecTransformer
Input [4]: [w_warehouse_name#12, i_item_id#7, sum#17, sum#18]
Keys [2]: [w_warehouse_name#12, i_item_id#7]
Functions [2]: [sum(CASE WHEN (d_date#10 < 2000-03-11) THEN inv_quantity_on_hand#3 ELSE 0 END), sum(CASE WHEN (d_date#10 >= 2000-03-11) THEN inv_quantity_on_hand#3 ELSE 0 END)]
Aggregate Attributes [2]: [sum(CASE WHEN (d_date#10 < 2000-03-11) THEN inv_quantity_on_hand#3 ELSE 0 END)#20, sum(CASE WHEN (d_date#10 >= 2000-03-11) THEN inv_quantity_on_hand#3 ELSE 0 END)#21]
Results [4]: [w_warehouse_name#12, i_item_id#7, sum(CASE WHEN (d_date#10 < 2000-03-11) THEN inv_quantity_on_hand#3 ELSE 0 END)#20 AS inv_before#22, sum(CASE WHEN (d_date#10 >= 2000-03-11) THEN inv_quantity_on_hand#3 ELSE 0 END)#21 AS inv_after#23]

(33) FilterExecTransformer
Input [4]: [w_warehouse_name#12, i_item_id#7, inv_before#22, inv_after#23]
Arguments: (CASE WHEN (inv_before#22 > 0) THEN ((cast(inv_after#23 as double) / cast(inv_before#22 as double)) >= 0.666667) END AND CASE WHEN (inv_before#22 > 0) THEN ((cast(inv_after#23 as double) / cast(inv_before#22 as double)) <= 1.5) END)

(34) WholeStageCodegenTransformer (6)
Input [4]: [w_warehouse_name#12, i_item_id#7, inv_before#22, inv_after#23]
Arguments: false

(35) TakeOrderedAndProjectExecTransformer
Input [4]: [w_warehouse_name#12, i_item_id#7, inv_before#22, inv_after#23]
Arguments: 100, [w_warehouse_name#12 ASC NULLS FIRST, i_item_id#7 ASC NULLS FIRST], [w_warehouse_name#12, i_item_id#7, inv_before#22, inv_after#23], 0

(36) VeloxColumnarToRow
Input [4]: [w_warehouse_name#12, i_item_id#7, inv_before#22, inv_after#23]

===== Subqueries =====

Subquery:1 Hosting operator id = 1 Hosting Expression = inv_date_sk#4 IN dynamicpruning#5
ColumnarBroadcastExchange (40)
+- ^ FilterExecTransformer (38)
   +- ^ FileSourceScanExecTransformer parquet spark_catalog.default.date_dim (37)


(37) FileSourceScanExecTransformer parquet spark_catalog.default.date_dim
Output [2]: [d_date_sk#9, d_date#10]
Batched: true
Location: InMemoryFileIndex [{warehouse_dir}/date_dim]
PushedFilters: [IsNotNull(d_date), GreaterThanOrEqual(d_date,2000-02-10), LessThanOrEqual(d_date,2000-04-10), IsNotNull(d_date_sk)]
ReadSchema: struct<d_date_sk:int,d_date:date>

(38) FilterExecTransformer
Input [2]: [d_date_sk#9, d_date#10]
Arguments: (((isnotnull(d_date#10) AND (d_date#10 >= 2000-02-10)) AND (d_date#10 <= 2000-04-10)) AND isnotnull(d_date_sk#9))

(39) WholeStageCodegenTransformer (1)
Input [2]: [d_date_sk#9, d_date#10]
Arguments: false

(40) ColumnarBroadcastExchange
Input [2]: [d_date_sk#9, d_date#10]
Arguments: HashedRelationBroadcastMode(List(cast(input[0, int, false] as bigint)),false), [plan_id=4]


